Approximate Canonical Correlation Analysis by Fuzzy Interval Data

Abstract

When the data is observed in some systems which involve human beings, there are cases that the data is not always exact or it is difficult to observe the data exactly. In these cases, if we can observe the data including the human vagueness then its observation becomes easy. Then, in this paper, on the assumption that the observations include the human vagueness, we treat the fuzzy interval data including unclear border of intervals. And, we give the processing method of the canonical correlation analysis by fuzzy interval data. But, it is difficalt to treat the fuzzy interval data exactly. So, in our method, we execute the canonical correlation analysis apploximately. And, the usefulness of our method is illustrated using computer simulation.